<p>With the growing adoption of automated guided vehicles (AGVs) in various industries, the integrated production and transportation scheduling problem (IPTSP) has emerged as a critical research focus. The IPTSP is classified as a strongly NP-hard problem due to the simultaneous scheduling of two resources: machines and transportation equipment. Meta-heuristic algorithms are one of the most popular and effective approaches to solving this problem. However, their effectiveness heavily depends on the choice of solution representation, which influences both the algorithm’s search space and convergence speed. This paper reviews the existing encoding and decoding methods and proposes a novel active decoding approach. Based on different combinations of encoding and decoding methods, six solution representations are identified, among which the newly proposed representation offers a trade-off between the search space and the algorithm’s efficiency. Specifically, four scenarios of IPTSP under different assumptions are first analyzed. Next, the variations in the six solution representations across unused scenarios and different layouts, as well as their respective encoding spaces and qualities, are summarized. Subsequently, the search efficiency of the six solution representations is evaluated using a genetic algorithm to analyze their performance under different scenarios, layouts, time ratios, and number of AGVs. Finally, the advantages, disadvantages and applicable scenes for each solution representation are summarized based on the experimental results and analysis. These findings provide valuable insights for designing more efficient algorithms to address the IPTSP.</p>